Intel Discloses Detailed Skylake Architecture Enhancements

MojoKid writes: Intel is still keeping a number of details regarding its complete Skylake microarchitecture and product line-up under wraps for a few more weeks, but at a public session at IDF, some of the design updates introduced with Skylake were detailed. Virtually every aspect of Skylake has been improved versus the previous-gen Haswell microarchitecture. I/O, Ring Bus, and LLC throughput has been increased, the graphics architecture has been updated to support DX12 and new eDRAM configurations, it has an integrated camera ISP, support for faster DDR4 memory, and more flexible overclocking features. All of these things culminate in a processor that offers higher IPC performance and improved power efficiency. There are also new security technologies dubbed Intel Software Guard Extensions (Intel SGX) onboard Skylake, which support new instructions to create and isolate enclaves from malware and privileged software attack, along with Memory Protection Extensions (Intel MPX) to help protect stack and heap buffer boundaries as well. A new technology, dubbed Intel Speed Shift, also allows Skylake to switch power states faster than previous-gen products, controlling P states fully in hardware, whereas previous-gen products required OS control. The end result is that Skylake can switch P states in 1ms, whereas it takes roughly 30ms with older processors.

Comments

Motherboard compatibility?

[Anonymous Coward]

I recently bought a socket 2011-v3 motherboard and put a "low-end" 6-core CPU on it. I wonder if this Skylake can be considered an upgrade path..

Re:Motherboard compatibility?

[Rockoon]

I wonder if this Skylake can be considered an upgrade path..

That would violate Intels strategy, which is never allow a realistic upgrade path.

Re: Motherboard compatibility?

[Anonymous Coward]

No it's not. You'll need to wait for Skylake-E or whatever they will call it. IMO it won't be worth it.

Re:Motherboard compatibility?

[SuricouRaven]

It's socket LGA 1151. New motherboard likely.

Re: Motherboard compatibility?

[myrdos2]

You'll need to wait for Skylake-E or whatever they will call it.

Sheesh. Skynet. They're going to call it Skynet.

Re:Motherboard compatibility?

[Anonymous Coward]

The new cpu changed the DMI link so that may be why.

Re:Motherboard compatibility?

[Anonymous Coward]

You've been smoking too much AMD, son. Intel requires new motherboards for everything.

Re:Motherboard compatibility?

[FreonTrip]

Necessary, sadly. They moved the VRM back out of the die, which necessitates a socket change. Naturally the team working on what's supposed to be its successor was the one who moved into the die in the first place, so they intend to put it [i]back[/i] for their revision. No such thing as reusing a perfectly good motherboard in Intel country.

Re:Motherboard compatibility?

[Anonymous Coward]

To be fair, Intel violates reality's strategy. If you even think your old mobo might be compatible with the new stuff, then you don't even need the new stuff. There is no reason anyone with socket 2011 stuff should be considering "upgrading." Even if you could do it, you wouldn't want to because you'd just be burning money for nearly imperceptible gain.

Re:Motherboard compatibility?

[eth1]

I wonder if this Skylake can be considered an upgrade path..

That would violate Intels strategy, which is never allow a realistic upgrade path.

I'm not even entirely sure you *need* one any more. I'm writing this on a first-gen i7-920 - not overclocked - I put together in 2009. It's STILL "fast enough" for pretty much anything, even without a GPU upgrade in the past three years. And I use it primarily for games (the only one that utterly crushed it was ARK:SE).

I am planning to upgrade within the next few months, now that Skylake and Win 10 are here, but the primary driver for that is moving from spinning disks to a good NVMe-based system, not because the CPU is too slow.

If you're careful building your system, they'll last long enough that virtually the entire system is outdated, anyway (except for stuff like the case and power supply, which you can usually re-use), so updating just one component isn't really worth doing.

BMI/BMI2

[Anonymous Coward]

So, did they really take away BMI/BMI2?

Re:BMI/BMI2

[Rockoon]

Sure looks like it

From the errata:

Executing CPUID with EAX = 7 and ECX = 0 may return EBX with bits [3] and [8] set, incorrectly indicating the presence of BMI1 and BMI2 instruction set extensions.

Attempting to use instructions from the BMI1 or BMI2 instruction set extensions will result in a #UD exception.

and in the errata summary, its currently labeled NO FIX so they dont even have a fix that will trap the exception and emulate the instructions (which would perform terribly anyways... but hey, working is better than not working.)

Re:BMI/BMI2

[Anonymous Coward]

I'm most concerned about the "may" part of that.

Didn't we just have this posted yesterday?

[Anonymous Coward]

Yesterday's news, news today!

Between CPUs & SSD improvements?

[Anonymous Coward]

It's nice to know that the next system I build (once every 3-5 yrs. typically) is going to be even better/faster/more efficient than the one I have currently.

* Gotta love electrical engineers etc. who build, design, prototype, & bring this improved stuff into production - 3 cheers for them.

(Good job Intel!)

APK

P.S.=> Typically & for example: I tend to get systems that are literally 2-4x as fast (in various areas, mostly CPU or Video performance usually - RAM lags a bit, but still improves too (they toss on more levels of L1/L2/L3/L4 caches for the CPU to make up for it imo)) since I purchase once every 3-5 yrs. so, it makes the money spent worth it imo @ least... apk

No G.P.P., no deal.

[jpellino]

They've been talking about this since 1978 and here we are no closer.

Re:No G.P.P., no deal.

[Anonymous Coward]

What computing purpose can you not use it for?

Sorry Intel, not enough

[JoeyRox]

Sounds like the last few generations - lots of incremental improvements and excellent technology but wont amount to much of a difference in general performance.

Re:Sorry Intel, not enough

[Anonymous Coward]

I tend to agree.
Where are the 32core chips for consumers?
What? They cost more than a thousand bucks?
WTF!

In the immortal words of a former wimbledon Champion

"You can't be serious!"

Re:Sorry Intel, not enough

[Kjella]

Sounds like the last few generations - lots of incremental improvements and excellent technology but wont amount to much of a difference in general performance.

Actually they've made quite substantial improvements, but Intel is using that to deliver 105% of the last generation's performance with far less resources. A 32nm Sandy Bridge (i7-2700k) is 216 mm^2, a 14nm Skylake (i7-6700k) is 122.4 mm^2. So on the same wafer Intel can produce 75% more processors. By letting AMD pave the way with APUs they've force-bundled integrated graphics killing low-end discrete chips without an antitrust whimper, almost 20% on Steam now game on Intel and Skylake adds 25% more shaders with 20 vs 16 EUs on regular desktop chips.

Perhaps the most important part of the Skylake announcement has gone largely unnoticed because we haven't seen it in any actual product yet is that Skylake will go up to 72 EUs, as opposed to 40/48 on Haswell/Broadwell. Since Broadwell quad core chips only launched a few months ago, the real comparison is Haswell which means a 72/40 = 80% increase in shaders, clearly Intel is planning to take the midrange laptop graphics too. A fully stacked Skywell seems to be a nVidia 950m-class competitor, both around 1100 GFLOPS and 26 times more powerful than "Intel HD Graphics" from five years ago.

Basically, we're being used. I think Intel knows as well as we do that no matter how fast they released processors people aren't going to throw away their three year old computer anymore. They'll sell new processors when the old go out of service or the market expands, not because it's outdated. So they're using their strengths for profit and market gain, because what's your high end alternative? It's either Intel or an extremely old FX processor or a severely under-powered laptop chip. They know your business is not going anywhere, it's only a matter of how long they need to wait.

Re:Sorry Intel, not enough

[thegarbz]

Depends on who's performance you're talking about. As a user of intense image editing software I would welcome more, bigger, better and faster and look forward to upgrading my several generations old system to this.

As a gamer I couldn't care less.

Huh?

[Anonymous Coward]

WTF R U talking about?

Slower in games, faster in vector maths

[Anonymous Coward]

If you forget the PAID Intel puff piece, the reality is that Skylake is a MAJOR step backwards over Broadwell (the Intel CPU arch between Haswell and Skylake) in certain applications, especially gaming. Meanwhile Skylake significantly beefed up its vector processing performance (and GPU).

I a way, the major loss of performance in certain games proves Skylake is quite a modified CORE architecture. But given the massive number of punters who buy high-priced i5 and i7 CPUs for PC gaming, this design choice by Intel is puzzling. DDR4 brings NOTHING but higher memory prices per GB. Luckily for Intel, Broadwell never got much traction on the desktop, and Skylake looks better against Haswell.

HOWEVER, what gamers want is a decent priced (sub 200 dollar) mainstream i5 with SIX true cores. Since the i5 was first introduced, ARM based SoC parts have gone from ONE CORE to TWELVE CORES (and much faster cores) for the SAME chip price. Intel, meanwhile, has barely improved the core performance since Sandybridge, has shrunk the CPU core to a fraction of its original die size, and charges MORE per core than when it first introduced the architecture.

Intel's pricing (and refusal to offer 6-core mainstream parts) is a consequence of Intel's effective MONOPOLY in the x86 space. AMD's current CPU offerings are a BAD JOKE, offering around 50% per core of Intel's core performance. No serious PC gamer would opt for anything less than a true 4-core i5. AMD isn't even in the picture.

Re:Slower in games, faster in vector maths

[0123456]

Intel's pricing (and refusal to offer 6-core mainstream parts) is a consequence of Intel's effective MONOPOLY in the x86 space. AMD's current CPU offerings are a BAD JOKE, offering around 50% per core of Intel's core performance. No serious PC gamer would opt for anything less than a true 4-core i5. AMD isn't even in the picture.

So why did I pay less for my i7 a couple of years ago than I did when I bought my Pentium-4 back in the days when AMD was actually competitive?

Intel's current competition is ARM, not AMD.

Re:Slower in games, faster in vector maths

[FlyHelicopters]

HOWEVER, what gamers want is a decent priced (sub 200 dollar) mainstream i5 with SIX true cores.

6 isn't enough of a jump over 4...

Give me 8 true cores and 16 threads, remove the IGP which I don't need for such a CPU...

Yes, yes, I know, Xeon and Haswell-E, but the reality is that the "need" for 8 core chips won't really happen until more of them hit the desktop market, and what AMD sells as 8 core doesn't count.

Re:Slower in games, faster in vector maths

[Anonymous Coward]

My AMD system works just fine for games. Not sure what you're doing wrong.

Unless you're just trolling with the "no true Scotsman" argument.

Re:Slower in games, faster in vector maths

[oh_my_080980980]

LMOL 50% per core. Try looking at performance tests. Intel is nowhere near twice the performance; which is what you would want at a minimum for 3x the cost.

Re:Slower in games, faster in vector maths

[oh_my_080980980]

It's AMD not ARM. ARM is not competing in the desktop world. Hardly in the laptop either. Try cell phones and tablets.

Re:Slower in games, faster in vector maths

[oh_my_080980980]

If there was ever a true multi-threaded application AMD would take the prize. As such Intel dominates because of single threaded applications.

Re:Slower in games, faster in vector maths

[0123456]

It's AMD not ARM. ARM is not competing in the desktop world.

Nor is AMD.

The difference is that ARM could compete there, and replace the 90% of desktop systems that don't need much CPU power.

Re:Slower in games, faster in vector maths

[Anonymous Coward]

DDR4 brings faster transfers, lower power consumption, room for future improvement - Power is the big thing here, most important for mobile. Which is intel's biggest market now. For desktops its just the march of progress. In the past each transition - SDRAM to DDR, DDR to DDR2, etc etc etc the new version was on parity or slightly slower or slightly faster depending on case. New implementations get kinks worked out. Each new standard has the potential for huge improvements.

As for the flap about cores and arm CPUs- You have no idea what you're talking about kid. ARM chips are nice and simple mobile parts but single thread is still king. You can throw cores at a problem all day long but without a smart way to get them to communicate quickly most applications scale up pretty badly.

I repeat. For the end user single thread performance is king. The fastest arm core available is about 5x slower than the cheapest, lowest end, slowest "core" based intel CPU.

It really shows in the server arena too. All that hype about super-dense low end cores is going nowhere. Arm products are nowhere and Intel has released the Xenon-D because everyone said. "yeah there's lots of cores but they're useless because the single thread is slow as shit"

At the moment the average end user has nothing they can do with 6 cores. Its fluff. Worse, it's die that sits there sucking power and throwing off heat.

Re:Slower in games, faster in vector maths

[tirefire]

I still have a "vintage 2006" Core 2 Quad in my gaming desktop. It's one of the original quad core monsters. It's so old it's on 65nm tech and it actually has two processor dies on one package that communicate over FSB. It's one of the last Intel chips without an on-die memory controller... it certainly COULD be paired with nice fast DDR3, but my old motherboard has DDR2-800 memory, and I'm not even running in dual channel because I buy mismatched old ram for cheap from friends.

When I noticed my pc was getting a bit slow a few years ago, I put a $20 Hyper 212+ cooler on there and overclocked it +45% to 3.5 Ghz. Had to raise the Vcore all the way to 1.6V, enough to make people on overclocking forums wince, but the thing is _stable_ and has been for years. I did the math and this "105 watt" rated cpu is sucking down 191 watts now. Absolutely amazed at how cheap good CPU cooling is today.

Paired with a Radeon 7850 I picked up for $180 a few years ago, it's good for 40+ fps on anything but the most poorly optimized single-threaded games I play. Any Crysis, any STALKER (even with punishing community-made mods with extra pretty everywhere), World of Tanks, Deus Ex:HR... had a blast playing all of them! At this point, pushing and prodding my el cheapo system to get a good frame rate in my games is half the satisfaction of gaming to me :P

Every once in a while I spec out a new build and see if it's worth it to me to plunk down a few hundred for an upgrade. So far, it just hasn't been worth it. Modern games are pretty good about using 3-4 cores; video encodes (the only other demanding thing I do) are embarrassingly parallel, and now that I have an SSD in my pc it hardly ever has a "senior moment". If I were building a new pc right now I'd probably go with an AMD vishera build. Never mind the tech is a few years old... they're cheap, they're FULL-FEATURED, they're FULLY OVERCLOCKABLE, and AMD doesn't play dirty like Intel, who rigs compilers and sends out PR shills everywhere. Or maybe you think the solution to hot chips is to spend millions on shills to insult AMD rather than just make an appropriately sized heatsink/fan that would cost another dollar to ship to the customer.

I've probably used enough Intel hate buzzwords to summon Hairyfeet to this thread, haha. But the man speaks the truth. This "Intel rules AMD drools" sentiment among gamers makes me wonder how many have tried fueling up thier Honda Civics with 100-low-lead airport gas -_- MAKES IT GO FASTER!!

Re:Slower in games, faster in vector maths

[Anonymous Coward]

It's AMD not ARM. ARM is not competing in the desktop world.

Nor is AMD.

The difference is that ARM could compete there, and replace the 90% of desktop systems that don't need much CPU power.

Historically and hysterically stupid post.

Re:Slower in games, faster in vector maths

[FreonTrip]

If there was ever a true multi-threaded application AMD would take the prize. As such Intel dominates because of single threaded applications.

There are embarrassingly well-threaded applications where AMD does well. The x264 encoder does a fantastic job and hammers all 8 of the cores in my FX-8320 at >90% utilization, and it was cheerfully faster at that than the i5 3570K I used to keep around. But IPC does ultimately win out, and Haswell's AVX2 support is sufficient to let an i5 4690K generally pull out ahead of my FX. That's especially true on interlaced media, where the deinterlacer's essentially single-threaded and the rest of the chip's basically waiting for that single core to finish before tackling the rest of the workload. For most other uses it's somewhere around a Nehalem quad core: certainly fast enough for what I do, but the overall performance outside of niche applications isn't impressive in absolute terms. At least it took to undervolting well, and it's a friggin' behemoth for virtualization.

Re:Slower in games, faster in vector maths

[Anonymous Coward]

Intel dominates because it beats the pants off of AMD in BOTH single AND multicore & multithreaded performance -- along with price/performance and performance/watt metrics. the ONLY place that AMD beats Intel is price, and only when you ignore *everything else* and ONLY look at price.

Even what lead AMD has in integrated graphics is shrinking, and is totally irrelevant, as anyone for whom that truly matters is going to have discrete graphics anyway.

Sorry fanbois.

Re:Slower in games, faster in vector maths

[FlyHelicopters]

There are embarrassingly well-threaded applications where AMD does well. The x264 encoder does a fantastic job and hammers all 8 of the cores in my FX-8320 at >90% utilization, and it was cheerfully faster at that than the i5 3570K I used to keep around. But IPC does ultimately win out, and Haswell's AVX2 support is sufficient to let an i5 4690K generally pull out ahead of my FX.

Try that same task on a i7 4790k and see even more speed... that is exactly the task that hyperthreading is for.

Re:Slower in games, faster in vector maths

[FlyHelicopters]

If there was ever a true multi-threaded application AMD would take the prize. As such Intel dominates because of single threaded applications.

Actually, it still generally doesn't...

An Intel Core i7 generally is faster than a 8 core FX chip, even in really, really well threaded applications.

The Intel chip is SO MUCH better per core and with the help of hyperthreading, it still wins.

Of course, this is all a moot point, if you're serious about such work, you're at least on Haswell-E with 8 true cores and 16 threads, or you're on a Xeon and this isn't even a conversation worth having.

For pure speed, it is all Intel.

Re:Slower in games, faster in vector maths

[FreonTrip]

In fairness Piledriver did a tolerable job against Sandy Bridge... The problem is that Intel hasn't exactly stood still since 2012, and between three generations of minor but cumulative performance and power improvements and the platform updates that came with them, there's a huge difference even in the consumer market. AMD doesn't have an answer to Haswell-E, and Opterons have languished in the same three year old doldrums as their FX cousins. Zen will be a make or break proposition; they can't continue the way they have been.

Re:Slower in games, faster in vector maths

[FreonTrip]

Yep - nothing like extra logical threads to take maximum advantage of a high performance architecture.

Re:Slower in games, faster in vector maths

[0123456]

Historically and hysterically stupid post.

Yes, yours is. But what about mine?

Why would Intel not be charging more for a current i7 than they used to charge for a Pentium-4, if they had no competition? AMD has nothing in that market, and, even in lower-end markets, they barely compete on performance and are vastly more power-hungry.

If Intel started charging $1000 for an i3, the market would be flooded with ARM desktops by Christmas.

Re:Slower in games, faster in vector maths

[FlyHelicopters]

I agree with all of what you've said...

AMD has a problem in that I think they are serving an ever smaller part of the market, the middle...

People either now want low power draw with "enough" performance, or they want "give me all you've got" performance, while not being totally stupid about the power.

Keep in mind that Haswell-E draws 30% less max power than the top AMD FX chip does, while crushing it in performance. Granted, it costs 5 times as much, but frankly if you're buying a machine for 3 years of use and you are using it to make money, the $1,000 cost of the chip is trivial.

Even the 6 core Haswell-E for $380 or so completely runs off from the best AMD has to offer, for not much more money.

It is a shame, I remember the days of Thunderbird from AMD, heck I remember the days of the 386DX-40, when you could get 486DX-25 performance for half the price. Those were days when a processor upgrade often doubled your performance, today it seems the gains are much smaller.

If AMD was pushing Intel, we might see $300 8 core Skylake chips for sale.

Re:Slower in games, faster in vector maths

[plcurechax]

HOWEVER, what gamers want is a decent priced (sub 200 dollar) mainstream i5 with SIX true cores.

6 isn't enough of a jump over 4...

For most home / personal computing (including high end video games) diminishing returns kick in hard past 4 cores. The problem is that in the few cases where tasks can be easily subdivided so as to utilize more than 4 cores, the cores will normally be stuck waiting for memory updates which continues to lag (speed / throughput wise) behind processor compute ability at an increasingly large gap which spans orders of magnitude. Of course the only known way to speed DRAM is to utilize more power, which goes against the general IT development trends (greener computing, more capable mobile).

The processor to memory speed gap is one of the reasons why Intel is investing in novel memory technology (phase change memory, etc.). The recent XPoint memory announcement hinted at potential future usage as "page swap" memory, replacing virtual memory management swapping pages out to disk (mechnical or solid state).

I haven't read all of Intel's releases this week, but one area I'm interested in is seeing how eDRAM (embedded DRAM) aka Crystal Well technology is going to end up being available and utilized across the Skylake line. In memory intensive benchmarks eDRAM has already shown considerably improvement in memory constrained benchmarks in Broadwell mobile processors, wheere it acts as an additional level of cache.

Give me 8 true cores and 16 threads, remove the IGP which I don't need for such a CPU...

Most people don't utilize more than 2 cores for more than 25-33% of the time, so the market for consumer-oriented many core processors just isn't there. People who really need the performance already just buy a Xeon.

Intel's "hyper-threading technology" is one of the biggest disappointments in many years, I wish they would let the branding and feature set die in obscurity like it deserves (IMHO).

Yes, yes, I know, Xeon and Haswell-E, but the reality is that the "need" for 8 core chips won't really happen until more of them hit the desktop market, and what AMD sells as 8 core doesn't count.

Well 8-16 core processors have been around for what, a bit less than a decade now? They won't really happen in the consumer / desktop market, because the market isn't demanding it (with purchasing dollars, not wishful thinking). Look at the very modest take-up of the Haswell-E X99 (LGA2011) 6 to 8 core processors released last year (August-Sept 2014 IIRC).

I love fast computers, personally I have a 6 core i7-5930K, and the performance difference for most home/consumer applications is so trivial that I don't notice a difference over using a 4 core i7-4790K except for in parallel benchmarks.

Re:Slower in games, faster in vector maths

[Anonymous Coward]

Nah, you say that but what your really, really want is the latest NEC vector prosessor: 4 cores at 1 GHz, 4 DDR3 2000 memory controllers per core at a measly 420W of power use per processor and memory board.

Re:Slower in games, faster in vector maths

[Anonymous Coward]

2x performance at 3x the cost is not a ratio I would ever expect to see anywhere. Frequently 50% performance gains are a tenfold increase in price.

Re:Slower in games, faster in vector maths

[FlyHelicopters]

I love fast computers, personally I have a 6 core i7-5930K, and the performance difference for most home/consumer applications is so trivial that I don't notice a difference over using a 4 core i7-4790K except for in parallel benchmarks.

Of course you don't, because you're at the bleeding edge...

The same was true when the Q6600 (Quad Conroe) launched, it simply didn't make a difference over the E6600 (Dual Conroe) due to the programs we were running at the time.

Windows 10 has much more potential than Windows XP did. Between OneDrive keeping it all synced, Cortana always listening, and the 20 other things running in the background, a core or 2 is busy quite often.

I have many machines used for production and testing, and I can now tell the difference between a dual core and quad core for general application use. In fact, the Q6600 that I have for older machine testing is actually faster in many cases than more modern dual cores, due to having 4 true cores. This doesn't mean it gets work done faster, rather it is SNAPPIER, which is what a lot of people relate a computer to being faster.

There is little done on a desktop that needs super high end single core performance, other than some games. Rather what is needed is to always have a spare core so the responsivness of the machine is instant. When I say "hey Cortana", I expect an instant response. A half second lag is noticeable.

Re:Slower in games, faster in vector maths

[FlyHelicopters]

I haven't read all of Intel's releases this week, but one area I'm interested in is seeing how eDRAM (embedded DRAM) aka Crystal Well technology is going to end up being available and utilized across the Skylake line. In memory intensive benchmarks eDRAM has already shown considerably improvement in memory constrained benchmarks in Broadwell mobile processors, wheere it acts as an additional level of cache.

The lag in DDR memory hasn't improved in a long time. The speed is faster, once you get a transfer going, modern DDR3/4 memory is rocket fast. The challenge is that the speed to the first byte hasn't improved for a decade, largely due to the latency and distance of the memory from the CPU, among other things.

eDRAM is a welcome change and frankly I expect to see more such improvements in the future as they run into various walls of physics, ranging from the size of atoms to the speed of light.

Something that hasn't changed in a long time is the overall design of the machine. Conroe was a throwback in many ways to the Pentium III, it was Intel admiting that Netburst was a mistake, without them saying it of course.

Nehalem was a change for the better and you can see it in the improved memory throughput and better design, but it is time for the next move. Skylake, for all its improvements, is just a beefed up Nehalem CPU in many ways.

Windows has changed a lot in the past 10 years. I have looked at what Intel talked about at the IDF and they see a future that I tend to agree with. The Internet of Things will actually cause a need for more CPU power, not less, due to having a hundred things connected. When your main computer is remotely talking to 10 things at a time, you're going to want more cores.

Something that I think Intel should look at is a Big/Little configuration. This will take support from Microsoft, it has to be baked into Windows so Windows knows what KIND of cores you have, but frankly having 4 super power cores and maybe 4 low power cores makes more sense for the use of silicon and battery life.

Not all these CPUs go into a desktop, I'd like a notebook with a 10 hour battery life (because it really isn't 10 hours when you use it and it is 2 years old).

Of course, while we're at it, can we get the weight down? My main laptop is a nice Acer 17" machine that is too heavy to really be called "portable". But it is nice with a true quad i7, 1080p IPS display, dual drives (SSD and HDD), and backlit keyboard (please, can we have all devices have that now?).

I just wish it was half the weight it is now. :(

Disappointed: "security" is just getting worse

[Anonymous Coward]

They keep adding "security" features, but I don't think they get what security actually is. It's great for the NSA to be able easily tap in, but that isn't MY SECURITY. That's not real security. Give me the damn specs and code already. This is the reason I'm abandoning X86. It's one big security nightmare where I don't have control over my PC and can't know what its doing any more.

Re:Disappointed: "security" is just getting worse

[Anonymous Coward]

There are also new security technologies dubbed Intel Software Guard Extensions (Intel SGX) onboard Skylake, which support new instructions to create and isolate enclaves from malware and privileged software attack

Really? Holy irony; this is a system designed to protect malware (NSA backdoors, DRM, etc.) from the owner of the machine. "Privileged software", ie. the operating system of your choice, damn well ought to have full control over the machine, and the ability to "attack" unwanted malware.

SGX alone could possibly be used in a reasonable way, but with UEFI managing the boot environment, it has the perfect opportunity to set up "isolated enclaves" for factory installed rootkits that can not be removed or easily detected. How can a system with SGX and UEFI ever be considered trustworthy? Even reflecting upon all of Intel's anti-competitive and anti-consumer behavior, they have really outdone themselves here.

ARMv8 is looking great, and the competitive market appears set to deliver the inexpensive, low-power, many-core chips that people want. Single-threaded performance may suffer a little, but I'd happily sacrifice it for better efficiency and the many other features that Intel cripples in the pursuit of profit.

And how soon will the DRM crowd glom on to SGX?

[Anonymous Coward]

And how long will it take DRM proponents to convert & utilize this feature to block music, video, documents, etc?

"There are also new security technologies dubbed Intel Software Guard Extensions (Intel SGX) onboard Skylake, which support new instructions to create and isolate enclaves from malware and privileged software attack, along with Memory Protection Extensions (Intel MPX) to help protect stack and heap buffer boundaries as well."

Perhaps not. Perhaps it cannot be used this way. But they will do so, if they can...

LLC? ISP? WTF?

[Lost Race]

"Limited Liability Corporation" and "Internet Service Provider" don't make much sense, but then again I'm pretty far behind the times on CPU architecture. Who knows what coprocessors they're spending their insane transistor budgets on these days.

OK, "ISP" appears to mean "Image Signal Processor". "LLC" could mean "Last Level Cache" or "Logical Link Control". "Last Level Cache" makes more sense in context, though this is the first time I've seen that phrase. Usually cache levels are explicitly numbered (first, second, third, etc).

It looks like they spelled out everything else except "IPC" which is obviously(?) "Instructions Per Cycle".

Good job there, author, submitter, and editor!

Re:LLC? ISP? WTF?

[Anonymous Coward]

Maybe it is Inter Process Communication

Re:LLC? ISP? WTF?

[Bite The Pillow]

Fuck Slashdot, I'm taking my eyeballs and clicks to an aggregator that provides a way to learn more about something if I'm interested. Like this new hyperlink thing I've been hearing about.

Wow. Really? It's taken less than a generation

[jpellino]

for /.ers to lose their Douglas Adams references?